February 29 (UPI) — British start-up Exscientia says it has developed its first drug using artificial intelligence (AI) and will be clinically tested in humans. The drug can be used to treat obsessive-compulsive disorder, which took less than a year from the time concepts were presented and the trial was prepared. Human trials will begin in March, but do you dare to take drugs designed with AI software?
AI has considerable advantages in designing drugs. There are many potentially useful molecules in drug development, and it is almost impossible for all medical researchers in the world to test them manually. But by using different types of AI, computer systems can find and mine the utility of different molecules, compare them with different parameters, and screen out the most promising compounds faster than humans.
AI is certainly powerful, but some question whether the technology is reliable or trustworthy and question what role AI should play in areas such as our healthcare. In drug research, many have expressed concern that the technology could be overhyped and that AI’s findings may not be as ground-breaking as we would like to believe.
Andrew Hopkins, chief executive of Exscientia, believes AI means stoicing fewer compounds for testing and fewer experiments are needed in the search for new drugs. “Active learning (a DI subcategory known as machine learning) algorithms automatically place the compounds with the highest amount of information in the priority of experimental synthesis and testing, enabling systems to learn faster than humans,” says Hopkins. “
Of course, AI is used more than just trying to synthesize new compounds. The technology can also be mined through scientific research and patient data, and help find new benefits for existing drugs, in addition to a wide range of other applications. And this use may not be limited to the development of drugs: researchers have used AI to track the spread of the new coronavirus, and are being deployed to address the opioid crisis in the United States.
Hopkins went on to explain that his company was the first platform to produce a Drugs for AI, which will be tested in clinical trials. The final compound, called DSP-1181, is expected to last longer and be more effective than other existing drugs for obsessive-compulsive disorder, he said. Japan’s Sumitomo Dainippon Pharma, which owns the drug, will oversee its clinical development. The first phase of human trials will also be conducted in Japan, which will test the safety of the drug and the body’s response to it.
Although the development of this new drug may seem remarkable, there are still some health questions. AI can help us find new molecules, but the molecules they find may eventually be similar to those we’ve studied. This is a warning from Derek Lowe, a Novartis researcher and chemist who is working on drug discovery. On his pharmaceutical blog, Lowe explains that simply finding a potential compound does not guarantee that scientists really understand the biochemical nature of the disease they are trying to treat, or that the drug will actually work.
“The problem is that pre-clinical drug optimization is not a problem, ” Says Lowe. In my opinion, this project saves months at best by sending their compounds into the same black box shredder, as every drug project does when it enters human trials. “
At the same time, the development of AI-assisted drugs raises the question of how these new research methods should be viewed. What is the difference between a drugs designed by AI and a single human-developed drug in the long run? Who should set rules for the use of AI in drug research?
Like all AI applications, health authorities are trying to figure out how best to research and regulate these tools. While the U.S. Food and Drug Administration (FDA) would not comment on the specific new drug, FDA spokesman Jeremy Kahn said the agency is committed to maintaining public health standards while protecting innovation, and its Drug Assessment and Research Center is evaluating regulatory considerations that AI tools can raise.
“The full role of AI in drug development has yet to be confirmed, and given the range of tools and technologies covered by this umbrella term, stakeholders understand AI in different ways,” Kahn explained by email. It is important that the evidentiary standards required to support drug approval remain unchanged, regardless of the technological advances involved. At the same time, an Exscientia spokesman said the drug must meet all requirements for a phase-1 trial in Japan.
It’s important to remember that if AI-based drug development does work, Exscientia and other pharmaceutical companies will make a lot of cash. This is also the case, as the number of major biotech investors investing in the technology is falling. Large pharmaceutical companies are increasingly investing in AI, with investors including German drugmakers Evotec and Bristol-Myers Squibb, which are working with several pharmaceutical giants to develop new drugs, including Bayer and GlaxoSmithKline. GlaxoSmithKline).
So, in the next few months, if selected, you may have the opportunity to try the AI-designed drug in Japan. Still, the latest developments have made us more optimistic than ever about the future, and AI’s design of new drugs is about to become a reality. But will they really treat our disease better than artificial drugs? We can only wait and see! (From: Vox Author: Rebecca Heilweil Compilation: Netease Intelligence Participation: Small)